Mental models like the testing pyramid and testing trophy provide good high level ideas about different categories of automated testing - unit, integration, end-to-end. They help us understand the tradeoffs between them and the value of each.

In real life, sometimes testing feels a lot less organized than that. Instead of a pyramid, it's more like sea monster, trying to pull our ship down with its many weird tentacles. Not only that, it's doing this when we are focused on getting our ship from A to B. We don't want to be fighting off this sea monster, we want to be delivering quality software. We only want to think about testing in the context of how it helps us deliver quality software quickly, and with confidence.

In this talk we'll focus on front-end testing and a decision framework to help us "test what matters, where it matters" and outsmart the sea monster. We will use Cypress to show examples of end-to-end, component, unit, and API level tests and see how they work together.